I had missed bcrypt while looking, it would seem. Thank you for telling
me my options!
Regards,
- Philip B.
On 9/28/2015 4:54 PM, John Clements wrote:
On Sep 28, 2015, at 5:56 AM, Philip Blair <peblair...@gmail.com> wrote:
Hello everyone,
I am considering a little project which will involve calling some FFI functions
in some C++ code. Now, since the Racket FFI binds to C, I know that I'll need
to make a wrapper file, which is fine. My question: How would something like
this potentially be distributed (in a form usable by ffi/unsafe) with a
package? Most every FFI package I see on the list binds to libraries which are
installed separately as dependencies. Since these wrappers would be entirely
specific to this package, I would imagine that distributing them together would
make the most sense. Is this possible? If so, how?
There are (at least) two different approaches. One is to distribute compiled
versions of the stub files, and the other is to provide source and compile it
at setup time. For an example of the former, you might want to take a look at
my “portaudio” package (though there may be a better canonical reference). For
an example of the latter, I see that the “bcrypt” package compiles at setup.
The former can break because you don’t provide enough flavors of compiled file.
The latter can break because your client doesn’t have a compiler.
Let me know if you have any questions!
John Clements
Thank you for the help.
Regards,
- Philip B.
--
You received this message because you are subscribed to the Google Groups "Racket
Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to racket-users+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
--
You received this message because you are subscribed to the Google Groups "Racket
Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to racket-users+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
